Abstract Study question Should the cut-off values for follicle number and ovarian volume in PCOS be revised, and should they be adjusted based on age or BMI? Summary answer The cut-off for follicle number per ovary can be maintained (age specific, not BMI adjusted), while for ovarian volume, it should be lowered or omitted. What is known already The 2023 International Guideline for PCOS concluded that the quality of evidence for recommendations regarding the ultrasound diagnosis of PCOS (FNPO ≥ 20 and/or ovarian volume (OV) ≥10 mL, in one or both ovaries) is low. It advises to regularly revise thresholds with advancing ultrasound technology and to investigate the impact of age on diagnostic accuracy. Consideration of other biological factors such as BMI is also warranted. Study design, size, duration Retrospective cross-sectional study, including data from 2001-2023, obtained through standardized endocrinological screening at the Department of Reproductive Endocrinology in the Erasmus Medical Center (Rotterdam, The Netherlands). Women diagnosed with PCOS during the screening were eligible for inclusion in the PCOS group. Women who underwent the same screening, had regular menstrual cycles, exhibited normal hormonal values, and showed no endocrinological pathology were eligible for inclusion as controls. Participants/materials, setting, methods A total of 2492 women (15-50 years) with PCOS and 152 controls were included. PCOS was diagnosed according to the 2003 Rotterdam criteria and the 2018 International Guideline. Exclusion criteria included: age < 15 or > 50 years, presence of any cysts (>10 mm) in one/both ovaries, use of oral contraceptives, and missing ultrasound data. The ovarian ultrasound markers were stratified by age in 5-year intervals. Receiver Operating Characteristic (ROC) analyses were performed for optimal cut-offs. Main results and the role of chance In our cohort, most women with PCOS up to age 35 exhibit a TFC ≥ 20, followed by a decline in FNPO above the age of 35. In a subgroup of women aged 35-40 (n = 24 PCOS, n = 24 controls, matched by age, assessed with ultrasound > 8 MHz), we found optimal cut-offs of 14.5 follicles (Left FNPO) and 14 follicles (Right FNPO), with AUC values of 0.94 and 0.93. Median OV in every age group was below the 10 mL cut-off. In women aged 20-35 (n = 75 PCOS, 75 controls, ultrasound > 8 MHz), optimal cut-offs were 6.5 mL (Left OV) and 7.4 mL (Right OV), with AUC values of 0.70 and 0.78. For 35-40 years, cut-offs were 8.0 mL (Left OV) and 7.2 mL (Right OV), including 24 women PCOS and 24 age-matched controls, with AUC values of 0.68 and 0.62. In our cohort 19/2413 women with PCOS (0.8%) would not have received the diagnosis PCOS if OV was not considered as a marker for PCOM. No clinically significant correlation between BMI and FNPO or OV was observed. Limitations, reasons for caution The optimal cut-off values should be interpreted with caution due to the small sample sizes in the ROC analyses. The control group includes women with isolated PCOM. There is an ongoing debate about whether they should be excluded in such analyses. Wider implications of the findings We suggest lowering or omitting the OV cut-off for PCOM detection and establishing age-specific FNPO cut-offs. Future research on ovarian markers in PCOS, should prioritize women with isolated ovulatory dysfunction or hyperandrogenism, aligning with the 2023 International PCOS Guideline recommendation to exclusively investigate PCOM in this subgroup. Trial registration number not applicable